Messages in this thread | | | Date | Sun, 26 Mar 2000 23:25:47 -0800 (PST) | From | Andre Hedrick <> | Subject | Re: IDE DMA slower in 2.3.99pre3 than in 2.2.14 and in 2.0.36 |
| |
On Sun, 26 Mar 2000, Alessandro Suardi wrote:
> According to both hdparm and 'dd'. > > Results, in MB/sec, for hdparm: > > buffer cache reads buffered disk reads > 2.2.14 multcount=0 36.26 10.58 > 2.2.14 multcount=8 36.36 10.27 > > 2.3.99pre3 multcount=0 36.99 8.11 > 2.3.99pre3 multcount=8 36.89 8.86 > > 2.3.99pre3 idebus=66 37.21 8.13 > multcount=0 > 2.3.99pre3 idebus=66 36.78 8.65 > multcount=8
Your system can not do "idebus=66".........try "idebus=33" and repeat. You have doubled the clocking and the settings of the PIO base that DMA/UDMA extentions work off of most likely reduced.
> > More - fishing in my old e-mail I can see that in 2.0.36 I used to > be able to run dd if=/dev/hda of=/dev/null bs=4096 count=16384 in > > 0.04user 1.90system 0:06.42elapsed 30%CPU (0avgtext+0avgdata 0maxresident)k > 0inputs+0outputs (81major+11minor)pagefaults 0swaps > > while 2.3.99-pre3 idebus=66 multcount=8 only gets to > > 0.02user 2.34system 0:08.91elapsed 26%CPU (0avgtext+0avgdata 0maxresident)k > 0inputs+0outputs (105major+13minor)pagefaults 0swaps > > It seems we're far slower than we used to be... this is a Quantum > Fireball SE4.3A, QDI Titanium IB board, TX430 chipset @66Mhz, > AMD K6-200 with 64MB 10ns SDRAM. > > > Thanks for any input, > > --alessandro <asuardi@uninetcom.it> <asuardi@it.oracle.com> > > Linux 2.2.14/2.3.99-pre3 glibc-2.1.2-11 gcc-2.95.2 binutils-2.9.5.0.31 >
Andre Hedrick The Linux ATA/IDE guy
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.tux.org/lkml/
| |